Advancing resilience in infrastructure projects through machine learning-driven models

Aug 4, 2026

Traditional risk management improves infrastructure utility but often fails to capture complexity and uncertainty. Growing emphasis on resilience—especially the readiness dimension—supports earlier threat detection and prevention. Machine learning (ML) can strengthen resilience modelling, yet empirically validated neural-network approaches remain limited, restricting accuracy, reliability, and practical use. This study builds a neural-network resilience model optimized for training efficiency and predictive performance, using feature-importance methods to enhance accuracy and interpretability. Results rank influential factors, highlighting operational resilience as the key determinant of project success, and offer adaptable guidance for decision-making across contexts.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages25

Reading Time1 h 3 min

AccessFree to download

Introduction

Critical infrastructure and complexity

Interdependence and risk of cascading disruptions

Disruption definition and implications

PRODUCTION PLANNING & CONTROL [https://doi.org/10.1080/09537287.2025.2583300](https://doi.org/10.1080/09537287.2025.2583300) Advancing resilience in infrastructure projects through machine learning-driven models Udechukwu Ojiakoa, b,c, Tse Chiu Wongb, Craig John Smithb, Maxwell Chipulud , M. K. S. Al-Mhdawie,f, Babajide Oyewog, and Lawrence Obokohc aBailey College of Engineering & Technology, Indiana State University, Terre Haute, USA; bDepartment of Design, Manufacturing & Engineering Management, University of Strathclyde, Glasgow, UK; cJohannesburg Business School, University of Johannesburg, Johannesburg, South Africa; dThe Business School, Edinburgh Napier University, Edinburgh, UK; eSchool of Computing, Engineering & Digital Technologies, Teesside University, Middlesbrough, UK; fDepartment of Civil, Structural and Environmental Engineering, Trinity College Dublin, Dublin, Ireland; gEssex Business School, University of Essex, Colchester, UK ABSTRACT Traditional risk management enhances infrastructure utility but remains limited in addressing complexity and uncertainty. This has shifted attention towards resilience, particularly the readiness dimension, to improve early threat detection and prevention. Machine Learning (ML) offers opportunities to advance resilience modelling, yet empirically validated ML-enabled approaches, especially those using neural networks, are scarce, restricting accuracy, reliability, and applicability. This study develops a neural network-enabled resilience model optimized for training efficiency and predictive performance. By incorporating established feature importance techniques, the model improves accuracy, interpretability, and the identification of influential factors. The findings extend resilience typologies by ranking factor importance in critical infrastructure, highlighting ‘Operational resilience’ as the most significant determinant of project success. Practically, the model provides managers with clearer insights for decision-making, supporting earlier threat recognition and stronger disruption detection. The framework is adaptable across resilience contexts with appropriate industry-or platform-specific modifications. ARTICLE HISTORY Received 20 April 2025 Accepted 27 October 2025 KEYWORDS Machine learning; resilience; readiness; infrastructure; project success; modelling 1. Introduction In recent years, global demand has surged for complex, multi-stakeholder, and technologically advanced projects that deliver not only individual critical infrastructure assets and platforms (Roehrich et al. 2024; Zhang et al. 2024), but also systems that are highly interdependent and connected (Grafius, Varga, and Jude 2020; Mao and Li 2018) . Critical infrastructure is defined as ‘ … an asset, system or part thereof […] which is essential for the maintenance of vital societal functions, health, safety, security, economic or social well-being of people, and the disruption or destruction of which would have a significant impact […] as a result of the failure to maintain those functions’ (European Council 2008) . These assets may be either physical or virtual (Sarker 2024) and can be privately owned or publicly operated (Alkhaleel 2024; Dick et al. 2019) . Critical infrastructure relies on a global network of interconnected physical and digital interconnected systems to manage functions such as power distribution, water management, security, and energy efficiency. Examples include airports, logistics hubs, data centres, and power grids, which are particularly vulnerable during the transition from project completion to operational handover (Al-Mazrouie et al. 2021; de Almeida Rodrigues et al. 2024a; Zhang et al. 2024) . These systems are increasingly interdependent, amplifying exposure to risks that often transcend organizational and national boundaries (Al-Mhdawi et al. 2025; Chopra and Khanna 2015; Kumar et al. 2021; Liu, Ferrario, and Zio 2017; Liu et al. 2022; Milanovi

FAQs

Why is resilience modelling gaining attention over traditional risk management in infrastructure projects?

Traditional risk management strengthens utility but inadequately addresses complexity and uncertainty. Resilience—particularly readiness—enables earlier threat detection and prevention.

What gap does the study address regarding machine learning for resilience?

Empirically validated ML-enabled resilience approaches, especially neural-network methods, are still scarce. This limits accuracy, reliability, and applicability.

How does the proposed model improve interpretability and factor identification?

It incorporates established feature-importance techniques, improving accuracy, interpretability, and the identification of influential factors.

Which factor is identified as most significant for project success?

The findings extend resilience typologies by ranking factor importance and identify operational resilience as the most significant determinant of project success.

Aug 4, 2026

Upload by Clementine